chore: bump pipecat version and fix tests (#263)

* chore: bump pipecat version and fix tests

* chore: add github workflow to run tests

* fix: install reqirements.dev.txt in test script

* fix: fix api-test action

* feat: add integration test

* test: add integration tests

* test: add test for function call mute strategy

api/services/filesystem/__init__.py

@@ -1,9 +1,11 @@
 from .base import BaseFileSystem
 from .minio import MinioFileSystem
+from .null import NullFileSystem
 from .s3 import S3FileSystem
 
 __all__ = [
     "BaseFileSystem",
     "S3FileSystem",
     "MinioFileSystem",
+    "NullFileSystem",
 ]

api/services/filesystem/null.py

@@ -0,0 +1,50 @@
+from typing import Any, BinaryIO, Dict, NoReturn, Optional
+
+from .base import BaseFileSystem
+
+
+class NullFileSystem(BaseFileSystem):
+    """No-op filesystem used when storage is not configured (e.g. tests).
+
+    Every operation raises so that any test that exercises storage fails
+    loudly instead of silently succeeding against a stub.
+    """
+
+    def _fail(self, op: str) -> NoReturn:
+        raise RuntimeError(
+            f"NullFileSystem.{op} called — storage is not configured. "
+            "Set ENVIRONMENT to a non-test value or inject a real filesystem fixture."
+        )
+
+    async def acreate_file(self, file_path: str, content: BinaryIO) -> bool:
+        self._fail("acreate_file")
+
+    async def aupload_file(self, local_path: str, destination_path: str) -> bool:
+        self._fail("aupload_file")
+
+    async def aget_signed_url(
+        self,
+        file_path: str,
+        expiration: int = 3600,
+        force_inline: bool = False,
+        use_internal_endpoint: bool = False,
+    ) -> Optional[str]:
+        self._fail("aget_signed_url")
+
+    async def aget_file_metadata(self, file_path: str) -> Optional[Dict[str, Any]]:
+        self._fail("aget_file_metadata")
+
+    async def aget_presigned_put_url(
+        self,
+        file_path: str,
+        expiration: int = 900,
+        content_type: str = "text/csv",
+        max_size: int = 10_485_760,
+    ) -> Optional[str]:
+        self._fail("aget_presigned_put_url")
+
+    async def adownload_file(self, source_path: str, local_path: str) -> bool:
+        self._fail("adownload_file")
+
+    async def acopy_file(self, source_path: str, destination_path: str) -> bool:
+        self._fail("acopy_file")

api/services/looptalk/audio_streamer.py

@@ -9,7 +9,6 @@ import asyncio
 from typing import Dict, Set
 
 from loguru import logger
-
 from pipecat.audio.utils import mix_audio
 from pipecat.frames.frames import (
     Frame,

api/services/looptalk/core/pipeline_builder.py

@@ -3,6 +3,10 @@
 from typing import Any, Dict
 
 from loguru import logger
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.processors.aggregators.llm_response_universal import (
+    LLMContextAggregatorPair,
+)
 
 from api.db.db_client import DBClient
 from api.services.looptalk.audio_streamer import get_or_create_audio_streamer
@@ -23,10 +27,6 @@ from api.services.pipecat.service_factory import (
 from api.services.workflow.dto import ReactFlowDTO
 from api.services.workflow.pipecat_engine import PipecatEngine
 from api.services.workflow.workflow import WorkflowGraph
-from pipecat.pipeline.pipeline import Pipeline
-from pipecat.processors.aggregators.llm_response_universal import (
-    LLMContextAggregatorPair,
-)
 
 
 class LoopTalkPipelineBuilder:

api/services/looptalk/internal_serializer.py

@@ -7,7 +7,6 @@
 """Internal frame serializer for agent-to-agent communication."""
 
 from loguru import logger
-
 from pipecat.frames.frames import (
     Frame,
     InputAudioRawFrame,

api/services/looptalk/internal_transport.py

@@ -11,8 +11,6 @@ import time
 from typing import Dict, Optional, Tuple
 
 from loguru import logger
-
-from api.services.looptalk.internal_serializer import InternalFrameSerializer
 from pipecat.frames.frames import (
     CancelFrame,
     EndFrame,
@@ -29,6 +27,8 @@ from pipecat.transports.base_input import BaseInputTransport
 from pipecat.transports.base_output import BaseOutputTransport
 from pipecat.transports.base_transport import BaseTransport, TransportParams
 
+from api.services.looptalk.internal_serializer import InternalFrameSerializer
+
 
 class InternalInputTransport(BaseInputTransport):
     """Input side of internal transport for agent-to-agent communication."""

api/services/looptalk/orchestrator.py

@@ -6,6 +6,8 @@ from pathlib import Path
 from typing import Any, Dict, Optional
 
 from loguru import logger
+from pipecat.pipeline.task import PipelineTask
+from pipecat.utils.run_context import set_current_run_id
 
 from api.db.db_client import DBClient
 from api.services.looptalk.internal_transport import (
@@ -13,8 +15,6 @@ from api.services.looptalk.internal_transport import (
     InternalTransportManager,
 )
 from api.services.pipecat.transport_setup import create_internal_transport
-from pipecat.pipeline.task import PipelineTask
-from pipecat.utils.run_context import set_current_run_id
 
 from .core.pipeline_builder import LoopTalkPipelineBuilder
 from .core.recording_manager import RecordingManager

api/services/pipecat/event_handlers.py

@@ -188,7 +188,12 @@ def register_event_handlers(
                         await engine.llm.queue_frame(LLMContextFrame(engine.context))
                 else:
                     logger.debug("Playing text greeting via TTS")
-                    await task.queue_frame(TTSSpeakFrame(greeting_value))
+                    # append_to_context=True so the assistant aggregator commits
+                    # the greeting to the LLM context once TTS finishes; without
+                    # it the LLM would re-greet on its first generation.
+                    await task.queue_frame(
+                        TTSSpeakFrame(greeting_value, append_to_context=True)
+                    )
             else:
                 logger.debug(
                     "Both pipeline_started and client_connected received - triggering initial LLM generation"

api/services/pipecat/run_pipeline.py

@@ -429,7 +429,6 @@ async def _run_pipeline(
     engine.set_audio_config(audio_config)
 
     assistant_params = LLMAssistantAggregatorParams(
-        expect_stripped_words=True,
         correct_aggregation_callback=engine.create_aggregation_correction_callback(),
     )
 

api/services/smart_turn/app.py

@@ -21,9 +21,8 @@ from fastapi import (
     status,
 )
 from fastapi.websockets import WebSocketState
-from scipy.io import wavfile
-
 from pipecat.audio.turn.smart_turn.local_smart_turn_v2 import LocalSmartTurnAnalyzerV2
+from scipy.io import wavfile
 
 LOG_LEVEL = (
     logging.DEBUG

api/services/smart_turn/websocket_smart_turn.py

@@ -20,7 +20,6 @@ from typing import Any, Dict, Optional
 import numpy as np
 import websockets
 from loguru import logger
-
 from pipecat.audio.turn.smart_turn.base_smart_turn import (
     BaseSmartTurn,
     SmartTurnTimeoutException,

api/services/storage.py

@@ -2,6 +2,7 @@ from loguru import logger
 
 from api.constants import (
     ENABLE_AWS_S3,
+    ENVIRONMENT,
     MINIO_ACCESS_KEY,
     MINIO_BUCKET,
     MINIO_ENDPOINT,
@@ -11,9 +12,9 @@ from api.constants import (
     S3_BUCKET,
     S3_REGION,
 )
-from api.enums import StorageBackend
+from api.enums import Environment, StorageBackend
 
-from .filesystem import BaseFileSystem, MinioFileSystem, S3FileSystem
+from .filesystem import BaseFileSystem, MinioFileSystem, NullFileSystem, S3FileSystem
 
 
 def get_storage_for_backend(backend: str) -> BaseFileSystem:
@@ -73,12 +74,18 @@ def get_current_storage_backend() -> StorageBackend:
     return StorageBackend.get_current_backend()
 
 
-# Create a single storage instance at module load time
-_backend = StorageBackend.get_current_backend()
-logger.info(
-    f"Initializing storage backend: {_backend.name} (value: {_backend.value}, ENABLE_AWS_S3={ENABLE_AWS_S3})"
-)
-storage_fs = get_storage_for_backend(_backend.value)
+# Create a single storage instance at module load time.
+# In the test environment we skip the real backend so import doesn't require
+# MinIO/S3 to be reachable; tests that need storage must inject a real fs.
+if ENVIRONMENT == Environment.TEST.value:
+    logger.info("ENVIRONMENT=test — using NullFileSystem (no storage backend)")
+    storage_fs: BaseFileSystem = NullFileSystem()
+else:
+    _backend = StorageBackend.get_current_backend()
+    logger.info(
+        f"Initializing storage backend: {_backend.name} (value: {_backend.value}, ENABLE_AWS_S3={ENABLE_AWS_S3})"
+    )
+    storage_fs = get_storage_for_backend(_backend.value)
 
 
 # For backward compatibility, keep get_storage() function

api/services/telephony/providers/ari/strategies.py

@@ -6,7 +6,6 @@ This module contains the business logic for Asterisk ARI call operations.
 from typing import Any, Dict
 
 from loguru import logger
-
 from pipecat.serializers.call_strategies import HangupStrategy, TransferStrategy
 
 

api/services/telephony/providers/ari/transport.py

@@ -1,15 +1,15 @@
 """ARI (Asterisk) transport factory."""
 
 from fastapi import WebSocket
-
-from api.services.pipecat.audio_config import AudioConfig
-from api.services.pipecat.audio_mixer import build_audio_out_mixer
-from api.services.telephony.factory import load_credentials_for_transport
 from pipecat.transports.websocket.fastapi import (
     FastAPIWebsocketParams,
     FastAPIWebsocketTransport,
 )
 
+from api.services.pipecat.audio_config import AudioConfig
+from api.services.pipecat.audio_mixer import build_audio_out_mixer
+from api.services.telephony.factory import load_credentials_for_transport
+
 from .serializers import AsteriskFrameSerializer
 from .strategies import ARIBridgeSwapStrategy, ARIHangupStrategy
 

api/services/telephony/providers/cloudonix/routes.py

@@ -8,6 +8,7 @@ import json
 
 from fastapi import APIRouter, Request
 from loguru import logger
+from pipecat.utils.run_context import set_current_run_id
 
 from api.db import db_client
 from api.services.telephony.factory import get_telephony_provider_for_run
@@ -15,7 +16,6 @@ from api.services.telephony.status_processor import (
     StatusCallbackRequest,
     _process_status_update,
 )
-from pipecat.utils.run_context import set_current_run_id
 
 router = APIRouter()
 

api/services/telephony/providers/cloudonix/strategies.py

@@ -3,9 +3,9 @@
 from typing import Any, Dict
 
 from loguru import logger
+from pipecat.serializers.call_strategies import HangupStrategy
 
 from api.services.telephony.providers.cloudonix.provider import CLOUDONIX_API_BASE_URL
-from pipecat.serializers.call_strategies import HangupStrategy
 
 
 class CloudonixHangupStrategy(HangupStrategy):

api/services/telephony/providers/cloudonix/transport.py

@@ -1,15 +1,15 @@
 """Cloudonix transport factory."""
 
 from fastapi import WebSocket
-
-from api.services.pipecat.audio_config import AudioConfig
-from api.services.pipecat.audio_mixer import build_audio_out_mixer
-from api.services.telephony.factory import load_credentials_for_transport
 from pipecat.transports.websocket.fastapi import (
     FastAPIWebsocketParams,
     FastAPIWebsocketTransport,
 )
 
+from api.services.pipecat.audio_config import AudioConfig
+from api.services.pipecat.audio_mixer import build_audio_out_mixer
+from api.services.telephony.factory import load_credentials_for_transport
+
 from .serializers import CloudonixFrameSerializer
 from .strategies import CloudonixHangupStrategy
 

api/services/telephony/providers/plivo/routes.py

@@ -9,6 +9,7 @@ from typing import Optional
 
 from fastapi import APIRouter, Header, Request
 from loguru import logger
+from pipecat.utils.run_context import set_current_run_id
 from starlette.responses import HTMLResponse
 
 from api.db import db_client
@@ -18,7 +19,6 @@ from api.services.telephony.status_processor import (
     _process_status_update,
 )
 from api.utils.common import get_backend_endpoints
-from pipecat.utils.run_context import set_current_run_id
 
 router = APIRouter()
 

api/services/telephony/providers/plivo/transport.py

@@ -1,15 +1,15 @@
 """Plivo transport factory."""
 
 from fastapi import WebSocket
-
-from api.services.pipecat.audio_config import AudioConfig
-from api.services.pipecat.audio_mixer import build_audio_out_mixer
-from api.services.telephony.factory import load_credentials_for_transport
 from pipecat.transports.websocket.fastapi import (
     FastAPIWebsocketParams,
     FastAPIWebsocketTransport,
 )
 
+from api.services.pipecat.audio_config import AudioConfig
+from api.services.pipecat.audio_mixer import build_audio_out_mixer
+from api.services.telephony.factory import load_credentials_for_transport
+
 from .serializers import PlivoFrameSerializer
 
 

api/services/telephony/providers/telnyx/routes.py

@@ -8,6 +8,7 @@ import json
 
 from fastapi import APIRouter, Request
 from loguru import logger
+from pipecat.utils.run_context import set_current_run_id
 
 from api.db import db_client
 from api.services.telephony.factory import get_telephony_provider_for_run
@@ -16,7 +17,6 @@ from api.services.telephony.status_processor import (
     StatusCallbackRequest,
     _process_status_update,
 )
-from pipecat.utils.run_context import set_current_run_id
 
 router = APIRouter()
 

api/services/telephony/providers/telnyx/transport.py

@@ -1,15 +1,15 @@
 """Telnyx transport factory."""
 
 from fastapi import WebSocket
-
-from api.services.pipecat.audio_config import AudioConfig
-from api.services.pipecat.audio_mixer import build_audio_out_mixer
-from api.services.telephony.factory import load_credentials_for_transport
 from pipecat.transports.websocket.fastapi import (
     FastAPIWebsocketParams,
     FastAPIWebsocketTransport,
 )
 
+from api.services.pipecat.audio_config import AudioConfig
+from api.services.pipecat.audio_mixer import build_audio_out_mixer
+from api.services.telephony.factory import load_credentials_for_transport
+
 from .serializers import TelnyxFrameSerializer
 
 

api/services/telephony/providers/twilio/routes.py

@@ -9,6 +9,7 @@ from typing import Optional
 
 from fastapi import APIRouter, Header, Request
 from loguru import logger
+from pipecat.utils.run_context import set_current_run_id
 from starlette.responses import HTMLResponse
 
 from api.db import db_client
@@ -18,7 +19,6 @@ from api.services.telephony.status_processor import (
     _process_status_update,
 )
 from api.utils.common import get_backend_endpoints
-from pipecat.utils.run_context import set_current_run_id
 
 router = APIRouter()
 

api/services/telephony/providers/twilio/strategies.py

@@ -8,7 +8,6 @@ from typing import Any, Dict
 
 import aiohttp
 from loguru import logger
-
 from pipecat.serializers.call_strategies import HangupStrategy, TransferStrategy
 
 

api/services/telephony/providers/twilio/transport.py

@@ -1,15 +1,15 @@
 """Twilio transport factory."""
 
 from fastapi import WebSocket
-
-from api.services.pipecat.audio_config import AudioConfig
-from api.services.pipecat.audio_mixer import build_audio_out_mixer
-from api.services.telephony.factory import load_credentials_for_transport
 from pipecat.transports.websocket.fastapi import (
     FastAPIWebsocketParams,
     FastAPIWebsocketTransport,
 )
 
+from api.services.pipecat.audio_config import AudioConfig
+from api.services.pipecat.audio_mixer import build_audio_out_mixer
+from api.services.telephony.factory import load_credentials_for_transport
+
 from .serializers import TwilioFrameSerializer
 from .strategies import TwilioConferenceStrategy, TwilioHangupStrategy
 

api/services/telephony/providers/vobiz/routes.py

@@ -10,6 +10,7 @@ from typing import Optional
 
 from fastapi import APIRouter, Header, Request
 from loguru import logger
+from pipecat.utils.run_context import set_current_run_id
 from starlette.responses import HTMLResponse
 
 from api.db import db_client
@@ -24,7 +25,6 @@ from api.utils.common import get_backend_endpoints
 from api.utils.telephony_helper import (
     parse_webhook_request,
 )
-from pipecat.utils.run_context import set_current_run_id
 
 router = APIRouter()
 

api/services/telephony/providers/vobiz/transport.py

@@ -7,15 +7,15 @@ Vobiz uses Plivo-compatible WebSocket protocol:
 
 from fastapi import WebSocket
 from loguru import logger
-
-from api.services.pipecat.audio_config import AudioConfig
-from api.services.pipecat.audio_mixer import build_audio_out_mixer
-from api.services.telephony.factory import load_credentials_for_transport
 from pipecat.transports.websocket.fastapi import (
     FastAPIWebsocketParams,
     FastAPIWebsocketTransport,
 )
 
+from api.services.pipecat.audio_config import AudioConfig
+from api.services.pipecat.audio_mixer import build_audio_out_mixer
+from api.services.telephony.factory import load_credentials_for_transport
+
 from .serializers import VobizFrameSerializer
 
 

api/services/telephony/providers/vonage/routes.py

@@ -9,6 +9,7 @@ from typing import Optional
 
 from fastapi import APIRouter, Request
 from loguru import logger
+from pipecat.utils.run_context import set_current_run_id
 
 from api.db import db_client
 from api.services.telephony.factory import get_telephony_provider_for_run
@@ -16,7 +17,6 @@ from api.services.telephony.status_processor import (
     StatusCallbackRequest,
     _process_status_update,
 )
-from pipecat.utils.run_context import set_current_run_id
 
 router = APIRouter()
 

api/services/telephony/providers/vonage/transport.py

@@ -1,13 +1,14 @@
 """Vonage transport factory."""
 
-from api.services.pipecat.audio_config import AudioConfig
-from api.services.pipecat.audio_mixer import build_audio_out_mixer
-from api.services.telephony.factory import load_credentials_for_transport
 from pipecat.transports.websocket.fastapi import (
     FastAPIWebsocketParams,
     FastAPIWebsocketTransport,
 )
 
+from api.services.pipecat.audio_config import AudioConfig
+from api.services.pipecat.audio_mixer import build_audio_out_mixer
+from api.services.telephony.factory import load_credentials_for_transport
+
 from .serializers import VonageFrameSerializer
 
 

api/services/workflow/disposition_mapper.py

@@ -1,14 +1,12 @@
 """Utility module for applying disposition code mapping."""
 
-from typing import Optional
-
 from loguru import logger
 
 from api.db import db_client
 from api.enums import OrganizationConfigurationKey
 
 
-async def apply_disposition_mapping(value: str, organization_id: Optional[int]) -> str:
+async def apply_disposition_mapping(value: str, organization_id: int | None) -> str:
     """Apply disposition code mapping if configured.
 
     Args:
@@ -46,32 +44,3 @@ async def apply_disposition_mapping(value: str, organization_id: Optional[int])
     except Exception as e:
         logger.error(f"Error applying disposition mapping: {e}")
         return value
-
-
-async def get_organization_id_from_workflow_run(
-    workflow_run_id: Optional[int],
-) -> Optional[int]:
-    """Get organization_id from workflow_run_id through the model relationships.
-
-    Args:
-        workflow_run_id: The workflow run ID
-
-    Returns:
-        The organization ID if found, otherwise None
-    """
-    if not workflow_run_id:
-        return None
-
-    try:
-        workflow_run = await db_client.get_workflow_run_by_id(workflow_run_id)
-        if not workflow_run or not workflow_run.workflow:
-            return None
-
-        workflow = workflow_run.workflow
-        if not workflow.user:
-            return None
-
-        return workflow.user.selected_organization_id
-    except Exception as e:
-        logger.error(f"Error getting organization_id from workflow_run: {e}")
-        return None

api/services/workflow/pipecat_engine.py

@@ -1,11 +1,5 @@
 from typing import TYPE_CHECKING, Awaitable, Callable, Optional, Union
 
-from api.services.pipecat.audio_playback import play_audio
-from api.services.workflow.disposition_mapper import (
-    apply_disposition_mapping,
-    get_organization_id_from_workflow_run,
-)
-from api.services.workflow.workflow import Node, WorkflowGraph
 from pipecat.adapters.schemas.tools_schema import ToolsSchema
 from pipecat.frames.frames import (
     BotStartedSpeakingFrame,
@@ -21,6 +15,11 @@ from pipecat.services.llm_service import FunctionCallParams
 from pipecat.services.settings import LLMSettings
 from pipecat.utils.enums import EndTaskReason
 
+from api.db import db_client
+from api.services.pipecat.audio_playback import play_audio
+from api.services.workflow.disposition_mapper import apply_disposition_mapping
+from api.services.workflow.workflow import Node, WorkflowGraph
+
 if TYPE_CHECKING:
     from pipecat.frames.frames import Frame
     from pipecat.services.anthropic.llm import AnthropicLLMService
@@ -114,6 +113,9 @@ class PipecatEngine:
         # Custom tool manager (initialized in initialize())
         self._custom_tool_manager: Optional[CustomToolManager] = None
 
+        # Cached organization ID (resolved lazily from workflow run)
+        self._organization_id: Optional[int] = None
+
         # Embeddings configuration (passed from run_pipeline.py)
         self._embeddings_api_key: Optional[str] = embeddings_api_key
         self._embeddings_model: Optional[str] = embeddings_model
@@ -141,10 +143,13 @@ class PipecatEngine:
 
     async def _get_organization_id(self) -> Optional[int]:
         """Get and cache the organization ID from workflow run."""
-        if self._custom_tool_manager:
-            return await self._custom_tool_manager.get_organization_id()
-        # Fallback for when manager is not yet initialized
-        return await get_organization_id_from_workflow_run(self._workflow_run_id)
+        if self._organization_id is None:
+            self._organization_id = (
+                await db_client.get_organization_id_by_workflow_run_id(
+                    self._workflow_run_id
+                )
+            )
+        return self._organization_id
 
     def _get_otel_context(self):
         """Extract the OTel Context from the task's TracingContext.
@@ -324,11 +329,7 @@ class PipecatEngine:
         )
 
         # Register function with LLM
-        self.llm.register_function(
-            name,
-            transition_func,
-            cancel_on_interruption=False,
-        )
+        self.llm.register_function(name, transition_func)
 
     async def _register_knowledge_base_function(
         self, document_uuids: list[str]

api/services/workflow/pipecat_engine_callbacks.py

@@ -14,7 +14,6 @@ import re
 from typing import TYPE_CHECKING
 
 from loguru import logger
-
 from pipecat.frames.frames import (
     LLMMessagesAppendFrame,
 )

api/services/workflow/pipecat_engine_context_summarizer.py

@@ -6,8 +6,6 @@ from typing import TYPE_CHECKING, Optional
 
 from loguru import logger
 from opentelemetry import trace
-
-from api.services.pipecat.tracing_config import ensure_tracing
 from pipecat.frames.frames import LLMContextSummaryRequestFrame
 from pipecat.utils.context.llm_context_summarization import (
     LLMContextSummarizationUtil,
@@ -15,6 +13,8 @@ from pipecat.utils.context.llm_context_summarization import (
 )
 from pipecat.utils.tracing.service_attributes import add_llm_span_attributes
 
+from api.services.pipecat.tracing_config import ensure_tracing
+
 if TYPE_CHECKING:
     from api.services.workflow.pipecat_engine import PipecatEngine
 

api/services/workflow/pipecat_engine_custom_tools.py

@@ -13,21 +13,6 @@ import uuid
 from typing import TYPE_CHECKING, Any, Dict, List, Optional
 
 from loguru import logger
-
-from api.db import db_client
-from api.enums import ToolCategory, WorkflowRunMode
-from api.services.pipecat.audio_playback import play_audio, play_audio_loop
-from api.services.telephony.call_transfer_manager import get_call_transfer_manager
-from api.services.telephony.factory import get_telephony_provider
-from api.services.telephony.transfer_event_protocol import TransferContext
-from api.services.workflow.disposition_mapper import (
-    get_organization_id_from_workflow_run,
-)
-from api.services.workflow.tools.calculator import get_calculator_tools, safe_calculator
-from api.services.workflow.tools.custom_tool import (
-    execute_http_tool,
-    tool_to_function_schema,
-)
 from pipecat.adapters.schemas.function_schema import FunctionSchema
 from pipecat.frames.frames import (
     FunctionCallResultProperties,
@@ -36,6 +21,18 @@ from pipecat.frames.frames import (
 from pipecat.services.llm_service import FunctionCallParams
 from pipecat.utils.enums import EndTaskReason
 
+from api.db import db_client
+from api.enums import ToolCategory, WorkflowRunMode
+from api.services.pipecat.audio_playback import play_audio, play_audio_loop
+from api.services.telephony.call_transfer_manager import get_call_transfer_manager
+from api.services.telephony.factory import get_telephony_provider
+from api.services.telephony.transfer_event_protocol import TransferContext
+from api.services.workflow.tools.calculator import get_calculator_tools, safe_calculator
+from api.services.workflow.tools.custom_tool import (
+    execute_http_tool,
+    tool_to_function_schema,
+)
+
 if TYPE_CHECKING:
     from api.services.workflow.pipecat_engine import PipecatEngine
 
@@ -75,7 +72,6 @@ class CustomToolManager:
 
     def __init__(self, engine: "PipecatEngine") -> None:
         self._engine = engine
-        self._organization_id: Optional[int] = None
 
     async def _play_config_message(
         self, config: dict, *, append_to_context: bool = False
@@ -122,12 +118,8 @@ class CustomToolManager:
         return False
 
     async def get_organization_id(self) -> Optional[int]:
-        """Get and cache the organization ID from workflow run."""
-        if self._organization_id is None:
-            self._organization_id = await get_organization_id_from_workflow_run(
-                self._engine._workflow_run_id
-            )
-        return self._organization_id
+        """Get the organization ID from the engine (shared cache)."""
+        return await self._engine._get_organization_id()
 
     async def get_tool_schemas(self, tool_uuids: list[str]) -> list[FunctionSchema]:
         """Fetch custom tools and convert them to function schemas.
@@ -215,13 +207,10 @@ class CustomToolManager:
                 function_name = schema["function"]["name"]
 
                 # Create and register the handler
-                handler, timeout_secs, cancel_on_interruption = self._create_handler(
-                    tool, function_name
-                )
+                handler, timeout_secs = self._create_handler(tool, function_name)
                 self._engine.llm.register_function(
                     function_name,
                     handler,
-                    cancel_on_interruption=cancel_on_interruption,
                     timeout_secs=timeout_secs,
                 )
 
@@ -244,19 +233,16 @@ class CustomToolManager:
             Async handler function for the tool
         """
         timeout_secs: Optional[float] = None
-        cancel_on_interruption = True
 
         if tool.category == ToolCategory.END_CALL.value:
-            cancel_on_interruption = False
             handler = self._create_end_call_handler(tool, function_name)
         elif tool.category == ToolCategory.TRANSFER_CALL.value:
             timeout_secs = 120.0
-            cancel_on_interruption = False
             handler = self._create_transfer_call_handler(tool, function_name)
         else:
             handler = self._create_http_tool_handler(tool, function_name)
 
-        return handler, timeout_secs, cancel_on_interruption
+        return handler, timeout_secs
 
     def _register_calculator_handler(self) -> None:
         """Register the built-in calculator function with the LLM."""
@@ -335,7 +321,7 @@ class CustomToolManager:
                     tool=tool,
                     arguments=function_call_params.arguments,
                     call_context_vars=self._engine._call_context_vars,
-                    organization_id=self._organization_id,
+                    organization_id=await self.get_organization_id(),
                 )
 
                 await function_call_params.result_callback(result)

api/services/workflow/pipecat_engine_variable_extractor.py

@@ -5,12 +5,12 @@ from typing import TYPE_CHECKING, Any, List
 
 from loguru import logger
 from opentelemetry import trace
+from pipecat.processors.aggregators.llm_context import LLMContext
+from pipecat.utils.tracing.service_attributes import add_llm_span_attributes
 
 from api.services.gen_ai.json_parser import parse_llm_json
 from api.services.pipecat.tracing_config import ensure_tracing
 from api.services.workflow.dto import ExtractionVariableDTO
-from pipecat.processors.aggregators.llm_context import LLMContext
-from pipecat.utils.tracing.service_attributes import add_llm_span_attributes
 
 if TYPE_CHECKING:
     from api.services.workflow.pipecat_engine import PipecatEngine

api/services/workflow/qa/analysis.py

@@ -4,6 +4,7 @@ import json
 from typing import Any
 
 from loguru import logger
+from pipecat.processors.aggregators.llm_context import LLMContext
 
 from api.db.models import WorkflowRunModel
 from api.services.gen_ai.json_parser import parse_llm_json
@@ -26,7 +27,6 @@ from api.services.workflow.qa.tracing import (
     setup_langfuse_parent_context,
 )
 from api.utils.template_renderer import render_template
-from pipecat.processors.aggregators.llm_context import LLMContext
 
 
 async def _run_llm_inference(

api/services/workflow/qa/node_summary.py

@@ -3,6 +3,7 @@
 from typing import Any
 
 from loguru import logger
+from pipecat.processors.aggregators.llm_context import LLMContext
 
 from api.db import db_client
 from api.db.models import WorkflowRunModel
@@ -10,7 +11,6 @@ from api.services.pipecat.service_factory import create_llm_service_from_provide
 from api.services.workflow.dto import NodeType, QANodeData
 from api.services.workflow.qa.llm_config import resolve_llm_config
 from api.services.workflow.qa.tracing import create_node_summary_trace
-from pipecat.processors.aggregators.llm_context import LLMContext
 
 NODE_SUMMARY_SYSTEM_PROMPT = (
     "You are analyzing a voice AI agent script. This is only a part of a larger script. "

api/services/workflow/qa/tracing.py

@@ -78,7 +78,6 @@ def add_qa_span_to_trace(
         return
     try:
         from opentelemetry import trace as otel_trace
-
         from pipecat.utils.tracing.service_attributes import add_llm_span_attributes
 
         tracer = otel_trace.get_tracer("pipecat")
@@ -122,9 +121,9 @@ def create_node_summary_trace(
     try:
         from opentelemetry import trace as otel_trace
         from opentelemetry.context import Context
+        from pipecat.utils.tracing.service_attributes import add_llm_span_attributes
 
         from api.services.pipecat.tracing_config import ensure_tracing
-        from pipecat.utils.tracing.service_attributes import add_llm_span_attributes
 
         if not ensure_tracing():
             return None